Dosage compensation in the mammal results in transcriptional silencing of one X-chromosome in the female sex. Known as X-chromosome inactivation (XCI), this process exemplifies epigenetic regulation and has served as a paradigm for understanding gene regulation by noncoding RNA (ncRNA) and chromatin-based mechanisms. XCI is controlled by the X-inactivation center (Xic), an X- linked region known for an abundance of genes that produce ncRNA. The initiation of XCI is regulated by two such ncRNA loci, Xist and its antisense partner, Tsix. Expression of Tsix blocks initiation of silencing, whereas expression of Xist guides chromosome-wide inactivation. While our understanding of how Tsix opposes Xist has continued to improve over the past 10 years, how Xist induces XCI remains largely unsolved. In the past year, my laboratory has made significant progress towards understanding how silencing might be induced: (i) We have identified the first Xist RNA-interacting factor as Polycomb repressive complex 2 (PRC2), the complex of proteins responsible for trimethylating H3-lysine27 (H3- K27me3). We have shown that Ezh2, the catalytic subunit of PRC2, is the specific RNA-binding subunit. Xist RNA directly binds this subunit via a repeated stem-loop structure at the 5' end of the RNA known as Repeat A. (ii) We have also discovered that Repeat A is an independent transcription unit, which we named RepA. The 1.6 kb RepA RNA initially recruits PRC2 to the X and is required for induction of full-length Xist RNA, which in turn binds PRC2 also and spreads H3-K27 methylation along the whole X. (iii) The antisense Tsix RNA binds PRC2 as well. We propose that competition between RepA and Tsix for PRC2 determines if and when XCI will initiate. (iv) Curiously, Dicer-deficient cells show defects in Xist localization, PRC2 recruitment, and H3-K27 methylation, suggesting that the RNAi pathway may also play a hand in Polycomb recruitment and function. Taken together, these data implicate ncRNA at the interface between Polycomb proteins and their target genes, and suggest that ncRNA may be a general mechanism by which chromatin modifers - which often lack sequence- specific DNA-binding subunits - may be directed to specific loci. This research proposal is aimed at understanding mechanisms by which Polycomb complexes are targeted to genetic loci by ncRNA. We will examine how RepA, Tsix, and PRC2 interact to trigger the initiation of XCI, address how silencing spreads along the X through Xist RNA and PRC2, and then determine whether ncRNA plays a role in targeting PRC2 to other genomic regions.